Scholarlink

<p>In the dairy industry, heat exchangers are essential for milk pasteurization, a process frequently compromised by fouling, which diminishes heat transfer efficiency and affects product quality. Addressing the challenges of fouling and subsequent cleaning is pivotal for ensuring the safety and operational efficiency of pasteurization processes. Our research introduces a comprehensive approach through the development of a predictive fouling model paired with a dynamic cleaning model, aimed at overcoming these obstacles. The predictive fouling model, which forecasts the separate masses of protein and mineral fouling on heat exchanger surfaces during milk pasteurization, synthesizes two distinct models previously proposed in the literature but never before integrated. This model integrates hydrodynamics, denaturation, and deposition kinetics, along with process characteristics to provide an accurate prediction. Additionally, a dynamic cleaning model was also established to track the transition of the initial deposit mass through an intermediate state before removal, enabling the optimization of cleaning times. This model considers key physico-chemical parameters such as temperature, concentration, and circulation rate of the cleaning agents, to predict and minimize cleaning times for both alkaline and acid solutions effectively. Preliminary validation studies with industrial partners confirmed the accuracy of the predicted fouling mass per area and the efficacy of cleaning with the proposed dynamic model. Pasteurization experiments at 80&#xa0;°C for 5&#xa0;h with a flow rate of 22.7m<sup>3</sup>/h demonstrated the model’s precision in estimating total deposit mass and the ratio of inorganic/organic components, consistent with the literature. This predicted deposit mass was further employed to revisit cleaning times for both acid and alkaline cleaning, and the efficacy of the predicted cleaning sequences has been successfully validated. This research marks a significant advancement in the predictive management of fouling and cleaning in milk pasteurization, offering both theoretical insights and practical solutions to enhance operational efficiency and product safety.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Predictive modeling for fouling and cleaning in heat exchangers during milk pasteurization

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年9月26日

查看全文
  • Deniz Yilmaz,
  • Khoang Kyong Nguen,
  • Mahieddine Chergui,
  • Heni Dallagi,
  • Guillaume Delaplace

摘要

In the dairy industry, heat exchangers are essential for milk pasteurization, a process frequently compromised by fouling, which diminishes heat transfer efficiency and affects product quality. Addressing the challenges of fouling and subsequent cleaning is pivotal for ensuring the safety and operational efficiency of pasteurization processes. Our research introduces a comprehensive approach through the development of a predictive fouling model paired with a dynamic cleaning model, aimed at overcoming these obstacles. The predictive fouling model, which forecasts the separate masses of protein and mineral fouling on heat exchanger surfaces during milk pasteurization, synthesizes two distinct models previously proposed in the literature but never before integrated. This model integrates hydrodynamics, denaturation, and deposition kinetics, along with process characteristics to provide an accurate prediction. Additionally, a dynamic cleaning model was also established to track the transition of the initial deposit mass through an intermediate state before removal, enabling the optimization of cleaning times. This model considers key physico-chemical parameters such as temperature, concentration, and circulation rate of the cleaning agents, to predict and minimize cleaning times for both alkaline and acid solutions effectively. Preliminary validation studies with industrial partners confirmed the accuracy of the predicted fouling mass per area and the efficacy of cleaning with the proposed dynamic model. Pasteurization experiments at 80 °C for 5 h with a flow rate of 22.7m3/h demonstrated the model’s precision in estimating total deposit mass and the ratio of inorganic/organic components, consistent with the literature. This predicted deposit mass was further employed to revisit cleaning times for both acid and alkaline cleaning, and the efficacy of the predicted cleaning sequences has been successfully validated. This research marks a significant advancement in the predictive management of fouling and cleaning in milk pasteurization, offering both theoretical insights and practical solutions to enhance operational efficiency and product safety.